netdev - [PATCH mlx5-next 07/14] net/mlx5: detect and enable bypass port select flow table

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed,  7 Sep 2022 16:36:29 -0700
From:   Saeed Mahameed <saeed@...nel.org>
To:     Saeed Mahameed <saeedm@...dia.com>,
        Leon Romanovsky <leonro@...dia.com>
Cc:     "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        Eric Dumazet <edumazet@...gle.com>, netdev@...r.kernel.org,
        Tariq Toukan <tariqt@...dia.com>,
        Jason Gunthorpe <jgg@...dia.com>, linux-rdma@...r.kernel.org,
        "Liu, Changcheng" <jerrliu@...dia.com>, Liu@...r.kernel.org,
        Mark Bloch <mbloch@...dia.com>
Subject: [PATCH mlx5-next 07/14] net/mlx5: detect and enable bypass port select flow table

From: "Liu, Changcheng" <jerrliu@...dia.com>

Use port selection capability port_select_flow_table_bypass
bit to detect and enable explicit port affinity even when
in lag hash mode.

Signed-off-by: Liu, Changcheng <jerrliu@...dia.com>
Reviewed-by: Mark Bloch <mbloch@...dia.com>
Reviewed-by: Saeed Mahameed <saeedm@...dia.com>
Signed-off-by: Saeed Mahameed <saeedm@...dia.com>
---
 .../net/ethernet/mellanox/mlx5/core/main.c    | 34 +++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index bec8d6d0b5f6..a284c97af213 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -652,6 +652,33 @@ static int handle_hca_cap_roce(struct mlx5_core_dev *dev, void *set_ctx)
 	return err;
 }
 
+static int handle_hca_cap_port_selection(struct mlx5_core_dev *dev,
+					 void *set_ctx)
+{
+	void *set_hca_cap;
+	int err;
+
+	if (!MLX5_CAP_GEN(dev, port_selection_cap))
+		return 0;
+
+	err = mlx5_core_get_caps(dev, MLX5_CAP_PORT_SELECTION);
+	if (err)
+		return err;
+
+	if (MLX5_CAP_PORT_SELECTION(dev, port_select_flow_table_bypass) ||
+	    !MLX5_CAP_PORT_SELECTION_MAX(dev, port_select_flow_table_bypass))
+		return 0;
+
+	set_hca_cap = MLX5_ADDR_OF(set_hca_cap_in, set_ctx, capability);
+	memcpy(set_hca_cap, dev->caps.hca[MLX5_CAP_PORT_SELECTION]->cur,
+	       MLX5_ST_SZ_BYTES(port_selection_cap));
+	MLX5_SET(port_selection_cap, set_hca_cap, port_select_flow_table_bypass, 1);
+
+	err = set_caps(dev, set_ctx, MLX5_SET_HCA_CAP_OP_MODE_PORT_SELECTION);
+
+	return err;
+}
+
 static int set_hca_cap(struct mlx5_core_dev *dev)
 {
 	int set_sz = MLX5_ST_SZ_BYTES(set_hca_cap_in);
@@ -696,6 +723,13 @@ static int set_hca_cap(struct mlx5_core_dev *dev)
 		goto out;
 	}
 
+	memset(set_ctx, 0, set_sz);
+	err = handle_hca_cap_port_selection(dev, set_ctx);
+	if (err) {
+		mlx5_core_err(dev, "handle_hca_cap_port_selection failed\n");
+		goto out;
+	}
+
 out:
 	kfree(set_ctx);
 	return err;
-- 
2.37.2